ID	原文	译文
3173	虽然浅层特征区分能力弱，但更忠于原始交易细节的描述，如何充分利用两者的优势是提升异常交易检测性能的关键，	Low-level features were more transaction content descriptive, although their distinguishing ability wasweaker than that of the high-level features. How to integrate them together to obtain complementary advantages was thekey to improve the detection performance.
3174	因此提出了特征融合方法自适应地桥接高层抽象特征与原始特征之间的鸿沟，自动去除其噪声和冗余信息，并挖掘两者的交叉特征信息获得最具区分力的特征。	Therefore, multi feature fusion methods were proposed to bridge the gap be-tween the two kinds of features. Moreover, these fusion methods can automatically remove the noise and redundant in-formation from the integrated features and further absorb the cross information, to acquire the most distinctive features.
3175	最后，结合以上方法提出区块链异常交易检测模型（BATDet），并通过Elliptic 数据集验证了所提模型在区块链异常交易检测领域的有效性。	Finally, block-chain abnormal transaction detection model (BATDet) was proposed based on the above presentedmethods, and its effectiveness in the abnormal transaction detection is verified.
3176	为了解决地面用户向多无人机边缘计算网络卸载数据时存在的地面被动窃听问题，提出了一种通过联合优化用户匹配和资源分配使系统能耗最小化的安全数据卸载策略。	To solve the problems of ground passive eavesdropping when ground users offload data to the mul-ti-UAV(unmanned aerial vehicle) edge computing network, a secure data offloading strategy that minimized system en-ergy consumption by jointly optimizing user matching and resource allocation was proposed.
3177	考虑了系统时延、通信资源、计算资源的限制，采用保密中断概率对数据卸载过程的安全性能进行约束。	Considering the constraintsof system delay, communication resources and computing resources, the probability of security interruption was used torestrict the security performance of the data offload process.
3178	利用块坐标下降和连续凸近似算法联合优化用户发送功率、卸载因子、无人机计算资源分配、干扰功率，并基于成对稳定的用户匹配算法最小化无人机系统总能耗。	By using block coordinate descent and successive convex approximation algorithm, the user transmission power, offload factor, UAV computing resource allocation and jammingpower were jointly optimized. A pair-wise stable user matching algorithm was proposed to minimize the total energyconsumption of UAV system.
3179	仿真数据表明，该算法可以实现数据的安全卸载，并且在能耗、时延等性能上优于传统策略。	Simulation results demonstrate that the algorithm can realize the safe offloading of data,and has good performance in energy consumption and delay by comparing with the conventional strategies.
3180	针对现有视距（LOS）定位方法在非视距（NLOS）环境中定位精度急剧恶化的问题，提出一种基于散射体信息的室内 NLOS 多站协作定位算法，可在完全没有 LOS 路径的情况下进行定位。	In indoor environments, the localization accuracy of existing line of sight (LOS) solutions will deteriorate se-verely in non-line-of-sight (NLOS) environment. In order to solve this problem, an scatterer information based indoorNLOS multiple base stations cooperative localization algorithm was proposed, which could realize localization when noLOS path was available.
3181	首先，利用多 AP 以及联合场景先验信息协同确定目标 NLOS 区域和散射体模糊区域；	Firstly, the target NLOS area and scatterer blur area were collaboratively determined throughmultiple AP and joint scene prior information.
3182	其次，根据信号的到达角对散射体区域进行约束，并在区域内搜索散射体的位置信息；	Secondly, the areas of scatterer were further constrained according to theangle of arrival.